Invest in Robots or in People? Amazon Makes the Lean Decision
I'm finally blogging about a Wall Street Journal article from just before Christmas – it's still a generally interesting topic, I think. The article was “Holiday Help: People vs. Robots“
It compares the labor and productivity strategies of two retailers – Crate & Barrel and Amazon. Amazon is known to be seriously embracing Lean principles, all the way up to the top and CEO Jeff Bezos. Check out this related video, via Pete Abilla and his Shmula blog, where Amazon Senior VP Marc Onetto gives a great lecture about Lean at Amazon.
You might recognize the robot as being a Toyota project. This is cool stuff, but it's being driven by necessity. With Toyota's aging population, there will be a shortage of caregivers for the elderly. With technology like this, Toyota often has its productivity credited to robots and automation. Those who know Toyota realize that Toyota factories tend to be relatively less automated than those of the “Detroit Three.” Toyota doesn't automate to get rid of factory labor at all costs – they tend to use automation where employee safety would be an issue or where quality is better with robots. It's not a technology-first mindset in Lean.
Some in the Lean world would bestow Toyota with a “luddite” reputation, with Toyota's focus on manual systems like kanban cards, heijunka boxes, and visual management boards. Toyota's not anti-technology. There is an effective middle ground, where Toyota uses technology smartly where it supports people and processes (see Principle 8 of the Toyota Way).
So back to the WSJ article.
Crate and Barrel's approach:
Zipping around the floor of a Crate & Barrel warehouse in Tracy, Calif., a cadre of 35 orange robots helps solitary human employees do the work of six people. The squat machines carry shelves with the company's 8,000 different products to people, who pick just the items they need to prepare online orders for shipment all without walking around the sprawling building. While Crate & Barrel sees orders quadruple during the holiday weeks, this warehouse gets by with just double the employees, about 90 humans, thanks to the robots.
Meanwhile, just outside of Phoenix, Amazon.com Inc. takes a more human-powered approach. There, employees walk 18 to 20 miles a day down aisles lined with shelves, filling library-style carts with the latest orders and carrying items back to packing stations. For the holiday season, this Phoenix facility one of about 20 that Amazon has across the U.S.quadruples its staff to about 1,200 to handle the holiday rush, with teams working 24 hours a day.
One advantage of people? They can participate in “kaizen” or continuous improvement. Robots cannot.
More from WSJ:
Amazon said it has found ways to make its human handpicking process more efficient through weekly team-brainstorming sessions, called “kaizen,” to remove waste from the process. For example, items ready to be picked are stored in shelves in a random manner so shelf space doesn't sit empty, and handheld computers help direct humans to walk the shortest possible path to pick items. Amazon has automated some parts of its fulfillment process, such as labeling boxes.
Being able to staff up and down, seasonally, provides a level of flexibility in cost and capacity, while automation might have a very fixed cost.
In Phoenix, the general manager of Amazon's fulfillment center, Prashant Bhat, said using humans, rather than robots or other automation, gives the company needed flexibility. “The variety of products that we have in this building is much greater than what you would see in a facility that has robots,” said Mr. Bhat.
Since retailing is a very seasonal business, some experts question the value of robots:
Dave Tavel, managing partner at AHS Inc., a firm that helps companies set up automated systems for their fulfillment centers, said he has looked at Kiva's robots but has yet to find an application for one of his clients where they made sense. One major concern, he said, is that it is expensive to buy sufficient robots for the busiest time of the year that sit largely idle other times.
The companies that make the automation and robotics, not surprisingly, argue that companies will be “held back” if they don't buy automation. OK, that's an expected argument.
Are the technology companies not Lean? One throws a curve ball by using classic Lean language:
“Kiva keeps humans in the loop, but we take out their nonvalue-added activity.”
Observing this from the outside, it seems Amazon's approach closely mirrors that of Toyota. Use technology where it makes sense. Have flexibility to scale labor up and down where you can't level load production.
I'm not saying robots or technology are bad. You have to use them smartly. I've seen many cases where hospitals made seemingly poor decisions with physical automation (in labs and pharmacies). That's going to be a topic for another blog post in the coming weeks…